Search

Abstract Background Hepatocellular carcinoma (HCC) is a common type of malignant tumor where the prognosis is dismal. Circular RNA (CircRNA) is a novel RNA that regulates downstream gene transcription and translation to influence the progression of HCC. However, the regulatory relationship that exists between E3 ligases, which is a class of post-translational modifying proteins, and circRNA remains unclear. Methods Based on the E3 ubiquitin ligase in the competitive endogenous RNA (ceRNA) network, a circRNA-regulated E3 ubiquitin ligase signature (CRE3UL) was developed. A CRE3UL signature was created using the least absolute shrinkage and selection operator (Lasso) and Cox regression analysis and merged it with clinicopathologic characteristics to generate a nomogram for prognosis prediction. The pRRophetic algorithm was utilized and immunological checkpoints were analyzed to compare the responses of patients in the high-risk group (HRG) and low-risk group (LRG) to targeted therapy and immunotherapy. Finally, experimental research will further elucidate the relationship between E3 ubiquitin ligase signature and HCC. Results HRG patients were found to have a worse prognosis than LRG patients. Furthermore, significant variations in prognosis were observed among different subgroups based on various clinical characteristics. The CRE3UL signature was identified as being an independent prognostic indicator. The nomogram that combined clinical characteristics and the CRE3UL signature was found to accurately predict the prognosis of HCC patients and demonstrated greater clinical utility than the current TNM staging approach. According to anticancer medication sensitivity predictions, the tumors of HRG patients were more responsive to gefitinib and nilotinib. From immune-checkpoint markers analysis, immunotherapy was identified as being more probable to assist those in the HRG. Conclusions We found a significant correlation between the CRE3UL signature and the tumor microenvironment, enabling precise prognosis prediction for HCC patients. Additionally, a nomogram was developed that performs well in predicting the overall survival (OS) of HCC patients. This provides valuable guidance for clinicians in devising specific personalized treatment strategies. Graphical Abstract

Abstract Induction of tumor vascular normalization is a crucial measure to enhance immunotherapy efficacy. cGAS-STING pathway is vital for anti-tumor immunity, but its role in tumor vasculature is unclear. Herein, using preclinical liver cancer models in Cgas/Sting-deficient male mice, we report that the interdependence between tumor cGAS and host STING mediates vascular normalization and anti-tumor immune response. Mechanistically, TET2 mediated IL-2/STAT5A signaling epigenetically upregulates tumor cGAS expression and produces cGAMP. Subsequently, cGAMP is transported via LRRC8C channels to activate STING in endothelial cells, enhancing recruitment and transendothelial migration of lymphocytes. In vivo studies in male mice also reveal that administration of vitamin C, a promising anti-cancer agent, stimulates TET2 activity, induces tumor vascular normalization and enhances the efficacy of anti-PD-L1 therapy alone or in combination with IL-2. Our findings elucidate a crosstalk between tumor and vascular endothelial cells in the tumor immune microenvironment, providing strategies to enhance the efficacy of combinational immunotherapy for liver cancer.

Three-dimensional reconstruction from point clouds is an important research topic in computer vision and computer graphics. However, the discrete nature, sparsity, and noise of the original point cloud contribute to the results of 3D surface generation based on global features often appearing jagged and lacking details, making it difficult to describe shape details accurately. We address the challenge of generating smooth and detailed 3D surfaces from point clouds. We propose an adaptive octree partitioning method to divide the global shape into local regions of different scales. An iterative loop method based on GRU is then used to extract features from local voxels and learn local smoothness and global shape priors. Finally, a moving least-squares approach is employed to generate the 3D surface. Experiments demonstrate that our method outperforms existing methods on benchmark datasets (ShapeNet dataset, ABC dataset, and Famous dataset). Ablation studies confirm the effectiveness of the adaptive octree partitioning and GRU modules.